This page contains a Flash digital edition of a book.
Henry F. Walling and the Mapping of New England’s Towns, 1849–1857

measured, recording bearings and distances between objects designated as boundary markers. This arrangement required a minimum of three men, one to handle the compass and two to manage the distance measurements. So long as no one moved the markers, metes-and-

bounds surveys worked well enough for surveying the bounds of small properties. Over larger distances such as the territory of a New England town, however, they tended to produce flawed results, due to the limitations of the instruments, the methods, and the surveyors themselves.7 So inaccurate were some of the Massachusetts town surveys of the 1830s that in 1844 Borden wrote to Walling “in one instance I placed the maps of two towns upon the space required to be filled by them, and found room enough left to place the map of Boston between them.”8 For all their flaws it appears that Walling used the boundaries delineated on the 1830s surveys as a framework over which other data could be superimposed. This view is supported by a comparison of several Walling maps with their 1830s counterparts, which reveals great similarities in their depictions of boundaries.9

Particularly useful in

this regard is his map of Lexington, Massachusetts (1853), which—unusually for a Walling map—gives the bearings and lengths of the town’s boundaries. These are quite similar and in some areas identical to those given on John Hales’s 1830 map of the town: clearly Walling used Hales’ data, updating it in places based on other sources or conceivably his own surveys. By contrast the interior details on the Hales and Walling maps vary substantially, indicating Walling must have obtained that data by other means. (See figures 5 and 6.) This pattern is repeated in other comparisons of Walling’s maps to their 1830s predecessors. Given the cost of paying two chainmen and a surveyor,

Walling would likely have repeated the boundary surveys only when the earlier work was hopelessly muddled or incomplete. Hence perhaps the qualifier on some of that maps that boundaries were derived “principally” or merely “in part” from existing surveys.

Odometer surveys

Walling took great care in depicting local roads on his town maps, and in this regard they generally appear far more exact than on the 1830s prototypes. In some cases, such as his maps of Andover and Concord (both 1852), he goes so far as to label each road in ½-mile or 1-mile intervals, indicating distances from the town center.

28 | The Portolan | Spring 2008

“Topographic Surveys of States,” an article published

by Walling toward the end of his life, provides strong albeit indirect evidence that Walling eschewed the traditional compass-and-chain method of road surveying. In a section discussing the production of his many large- scale county maps, he recalls that “the plan was adopted of traversing all the public highways by course and distance, using the ordinary surveyor’s compass for directions, and the revolutions of a wheel for distances.”10

The “wheel”

was an odometer, described by Ristow as “a small brass circular box that housed a series of cogwheels which regulated the motion of an index on a dial plate fixed to its exterior. The dial recorded the number of revolutions of a wheel attached to the box.”11

A surveyor would roll

the wheel along the path to be surveyed, then multiply the number of revolutions by the circumference of the wheel to determine the distance traveled. This data was complemented by compass readings to determine the bearing of each stretch of road. Given that the coverage of Walling’s town maps

overlaps that of his county maps, and that their periods of production overlapped, it seems reasonable that the former also drew heavily on odometer surveys. This method had two advantages: First, it was inexpensive, requiring but one man, whereas traditional compass- and-chain surveys required a minimum of three. Its second advantage, Walling argued, was a surprising level of accuracy: “By the system of traverses adopted, all the highways were surveyed and platted continuously in a network. Each closed circuit of this network not only checked itself, but served to check adjacent circuits. Usually the errors of closure did not exceed one or two per cent. of the distance traversed.”12

This accuracy also made

these odometer surveys a valuable means for checking existing boundary surveys, at least at points where roads crossed town lines. When preparing his county maps—and presumably

the town maps as well—Walling supplemented the quantitative data obtained by odometer surveys with qualitative data from simple observation: As he or his assistants traversed local roads, they made notes about the locations of dwellings and other structures, the points where waterways crossed, and the nature of the terrain. They then superimposed this information on the depiction of the road network: “Upon the final adjustment of the network of roads, the other details of topography, namely, the dwellings, streams, etc., were supplemented.”13

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16